IMAPS trial
From UMCN Radiology Research
IMAPS: International Multi-Centre Assessment of Prostate MR Spectroscopy
Document: Februari 2004
1 Background, Rationale and Objectives
1.1 Background Presently prostate cancer is the second most frequent cause of cancer related death in men. In the diagnosis of prostate cancer digital rectal examination (DRE) and especially the determination of the blood level of prostate specific antigen (PSA) are important first screening tools. If the findings are suspicious this may be followed by an investigation with transrectal ultrasound (TRUS). Subsequently, TRUS-guided multiple biopsy and histopathology can be performed to obtain a more definitive diagnosis. However, TRUS guidance may fail to detect cancer in the prostate. Other methods, including conventional T2 weighted MRI, are also not optimal in this respect. Proper non-invasive assessment of the presence of cancer tissue in the prostate, its location and possible heterogeneous extent is not only important for biopsy guidance but more recently also for treatments such as IMRT (Intensity Modulated Radio Therapy). Treatment selection depends on patient age and health, cancer stage and grade, morbidity and mortality of treatment, together with the preference of the patient and physician. The mainstay of treatment for organ confined disease is either radical prostatectomy or curative radiotherapy. These are only considered to be options if there is no spread to the seminal vesicles (seminal invasion), extension through the prostatic capsule (extracapsular extension) or metastatic disease (TNM stage £ T2N0M0). In staging of prostate cancer the detection of extraprostatic disease is thus an important clinical finding. In order to improve non-invasive detection of cancer in the prostate, its location, heterogeneous extent, grade and stage, currently several new MR techniques are being explored. Among these proton 3D Spectroscopic Imaging has shown promising results.
1.2 Rationale
While the contrast behaviour of adeno-carcinoma of the prostate compared to surrounding healthy prostate tissue appears to be ambiguous for many of the traditional diagnostic methods, analytical and clinical studies have shown significant differences if the metabolic state of the different tissue types is depicted by 1H MRS.
The major signals of 1H-MRS of the prostate are · the strongly coupled signal of citrate (Ci) at 2.60 ppm; · the signal of choline-containing compounds (Cho) at 3.20 ppm · the signal of creatine-containing compounds (Cr) at 3.04 ppm
Ci is produced by epithelial cells of the healthy prostate and is secreted in the luminal space. Cr-containing compounds are involved in the energy metabolism of cells and Cho-containing compounds are considered as an important precursor in cell membrane synthesis.
Preceding studies have shown the following behaviour of prostate metabolism as detected by 1H MRS: Healthy prostate tissue shows a higher signal of Ci than of Cho; the effect is more pronounced in the peripheral zone of the prostate and less pronounced in the central gland. Benign hyperplasia of the prostate and prostatis of the peripheral zone may show similar Cho/Ci ratios as healthy prostate tissue. The metabolic state of adeno-carcinomas occurring in the prostate appears to be markedly different: This tissue often shows an elevated choline-signal, while the citrate-signal is reduced or even missing.
This is the primary objective of the study, to prove that carcinoma tissue can be distinguished from other prostate tissue by considering only signal ratios obtained by 1H MRS. The gold standard for distinguishing the tissue types are the analysis of whole-mount sections of the resected prostate by a histo-pathologist. Anatomical MRI data is only required for the purpose of assigning spectra to certain anatomical areas.
Due to fact that performed prostatectomy is an inclusion criterium, the clinical validity of the study will be constrained to forms of carcinomas for which this therapy is to be preferred.
Additionally to the primary objectives, secondary study objectives aim at proving the accuracy of the diagnostic method of 1H MRS; since it is advantageous to pursue the secondary objectives by healthy volunteer exams, the validity of the study will be widened: It will be investigated whether it is possible to distinguish patients planned for prostatectomy from age-matched healthy volunteers by using 1H-MRS data.
An interesting aspect of MRS as a diagnostic method is that the result can be evaluated semi-automatically (= automatic with expert supervision). Given that the MRS data contains relevant clinical information, it could be used to increase the evaluation accuracy of less experienced readers, and thereby reduce inter-observer variability.
1.3 Objectives
Objective 1, Primary Objective Proving that 1H CSI data allows for detecting prostate carcinoma in the two major anatomic areas of the prostate, i.e. the peripheral zone and the central gland.
Objective 2, Primary Objective: Proving that 1H CSI data allows for localising prostate carcinoma. Objective 3, Secondary Objective: Documenting inter-subject variability of MRS exams
Objective 4, Secondary Objective: Documenting inter-site variability of MRS exams
Objective 5, Secondary Objective: Documenting intra-subject variability of MRS exams
Objective 6, Secondary Objective: Documenting age-related variances of metabolite signal ratios
Objective 7, Secondary Objective: Demonstrating the robustness of the MRS technique
2 Trial Organisation, Responsibilities
Principal investigator:
Arend Heerschap, UMCN, the Netherlands: a.heerschap@rad.umcn.nl responsibilities: · scientific advisor, speaker of the study group · organizer of the wrap-up meeting · organizer and correspondence author of scientific publications Second responsible person: Tom Scheenen Ph.D., t.scheenen@rad.umcn.nl
Responsible Clinical Advisor:
Jelle Barentsz, MD, Radiologist, UMCN, the Netherlands: j.barentez@rad.umcn.nl responsibilities: · advisor and speaker of the study group with respect to clinical issues
Responsible Biometrician: Robert Laheij, PhD, Medical Technology Assessment , UMCN , The Netherlands Responsibilities Advisor statistical aspects
Participating Institutions:
- Radboud University Nijmegen Medical Center, Netherlands
- Siemens Medical Solutions, Germany
- University Hospital, Katholieke Universiteit Leuven, Belgium
- Jules Bordet Institute, Brussels, Belgium
- The German Cancer Research Institute, DKFZ, Germany
- The University of Tuebingen, Germany
- Shanghai Changhai Hospital, China
- Loma Linda University Health Science Centre, USA
- Ghent University Hospital, Belgium
Responsibilities of the clinical institutions:
· risk assessment of study procedures, approval of study procedures e.g. by an independent ethical committee
· MR exams of study subjects;
participation requires contribution of at least 3 complete patient exams, containing CSI datasets for which 40% of the useful voxels are valid.
· histopathological examination
· data transfer to evaluation centre
· reviewing evaluation results as defined by the data evaluation plan
· reviewing protocol amendments
Supporting industrial partners:
Siemens Medical Solutions, Erlangen, Germany. short name: Siemens responsible person: Stefan Röll Ph.D., stefan.roell@siemens.com responsibilities: · initiating the trial, supporting trial organization · supporting participating sites technically during the trial · evaluation centre I
Medrad Inc., Pittsburgh, Pennsylvania, USA short name: Medrad site specifier in data sheets: n.a. responsible persons: Currie Crookston Ph.D., ccrookston@medrad.com Donna Feo, dfeo@medrad.com responsibilities: · providing participating sites with ER coils for trial exams · supporting participating sites with respect to coil issues
The participating clinical centres and supporting industrial partners are referred to as the study group.